summaryrefslogtreecommitdiff
path: root/fs/btrfs/raid56.h
blob: 91d5c0adad151edcb378e895d5381e7e44b7aab6 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
/* SPDX-License-Identifier: GPL-2.0 */
/*
 * Copyright (C) 2012 Fusion-io  All rights reserved.
 * Copyright (C) 2012 Intel Corp. All rights reserved.
 */

#ifndef BTRFS_RAID56_H
#define BTRFS_RAID56_H

#include <linux/workqueue.h>
#include "volumes.h"

enum btrfs_rbio_ops {
	BTRFS_RBIO_WRITE,
	BTRFS_RBIO_READ_REBUILD,
	BTRFS_RBIO_PARITY_SCRUB,
	BTRFS_RBIO_REBUILD_MISSING,
};

struct btrfs_raid_bio {
	struct btrfs_io_context *bioc;

	/*
	 * While we're doing RMW on a stripe we put it into a hash table so we
	 * can lock the stripe and merge more rbios into it.
	 */
	struct list_head hash_list;

	/* LRU list for the stripe cache */
	struct list_head stripe_cache;

	/* For scheduling work in the helper threads */
	struct work_struct work;

	/*
	 * bio_list and bio_list_lock are used to add more bios into the stripe
	 * in hopes of avoiding the full RMW
	 */
	struct bio_list bio_list;
	spinlock_t bio_list_lock;

	/*
	 * Also protected by the bio_list_lock, the plug list is used by the
	 * plugging code to collect partial bios while plugged.  The stripe
	 * locking code also uses it to hand off the stripe lock to the next
	 * pending IO.
	 */
	struct list_head plug_list;

	/* Flags that tell us if it is safe to merge with this bio. */
	unsigned long flags;

	/*
	 * Set if we're doing a parity rebuild for a read from higher up, which
	 * is handled differently from a parity rebuild as part of RMW.
	 */
	enum btrfs_rbio_ops operation;

	/* How many pages there are for the full stripe including P/Q */
	u16 nr_pages;

	/* How many sectors there are for the full stripe including P/Q */
	u16 nr_sectors;

	/* Number of data stripes (no p/q) */
	u8 nr_data;

	/* Numer of all stripes (including P/Q) */
	u8 real_stripes;

	/* How many pages there are for each stripe */
	u8 stripe_npages;

	/* How many sectors there are for each stripe */
	u8 stripe_nsectors;

	/* First bad stripe, -1 means no corruption */
	s8 faila;

	/* Second bad stripe (for RAID6 use) */
	s8 failb;

	/* Stripe number that we're scrubbing  */
	u8 scrubp;

	/*
	 * Size of all the bios in the bio_list.  This helps us decide if the
	 * rbio maps to a full stripe or not.
	 */
	int bio_list_bytes;

	refcount_t refs;

	atomic_t stripes_pending;

	atomic_t error;

	struct work_struct end_io_work;

	/* Bitmap to record which horizontal stripe has data */
	unsigned long dbitmap;

	/* Allocated with stripe_nsectors-many bits for finish_*() calls */
	unsigned long finish_pbitmap;

	/*
	 * These are two arrays of pointers.  We allocate the rbio big enough
	 * to hold them both and setup their locations when the rbio is
	 * allocated.
	 */

	/*
	 * Pointers to pages that we allocated for reading/writing stripes
	 * directly from the disk (including P/Q).
	 */
	struct page **stripe_pages;

	/* Pointers to the sectors in the bio_list, for faster lookup */
	struct sector_ptr *bio_sectors;

	/*
	 * For subpage support, we need to map each sector to above
	 * stripe_pages.
	 */
	struct sector_ptr *stripe_sectors;

	/* Allocated with real_stripes-many pointers for finish_*() calls */
	void **finish_pointers;
};

/*
 * For trace event usage only. Records useful debug info for each bio submitted
 * by RAID56 to each physical device.
 *
 * No matter signed or not, (-1) is always the one indicating we can not grab
 * the proper stripe number.
 */
struct raid56_bio_trace_info {
	u64 devid;

	/* The offset inside the stripe. (<= STRIPE_LEN) */
	u32 offset;

	/*
	 * Stripe number.
	 * 0 is the first data stripe, and nr_data for P stripe,
	 * nr_data + 1 for Q stripe.
	 * >= real_stripes for
	 */
	u8 stripe_nr;
};

static inline int nr_data_stripes(const struct map_lookup *map)
{
	return map->num_stripes - btrfs_nr_parity_stripes(map->type);
}

#define RAID5_P_STRIPE ((u64)-2)
#define RAID6_Q_STRIPE ((u64)-1)

#define is_parity_stripe(x) (((x) == RAID5_P_STRIPE) ||		\
			     ((x) == RAID6_Q_STRIPE))

struct btrfs_device;

void raid56_parity_recover(struct bio *bio, struct btrfs_io_context *bioc,
			   int mirror_num);
void raid56_parity_write(struct bio *bio, struct btrfs_io_context *bioc);

void raid56_add_scrub_pages(struct btrfs_raid_bio *rbio, struct page *page,
			    unsigned int pgoff, u64 logical);

struct btrfs_raid_bio *raid56_parity_alloc_scrub_rbio(struct bio *bio,
				struct btrfs_io_context *bioc,
				struct btrfs_device *scrub_dev,
				unsigned long *dbitmap, int stripe_nsectors);
void raid56_parity_submit_scrub_rbio(struct btrfs_raid_bio *rbio);

struct btrfs_raid_bio *
raid56_alloc_missing_rbio(struct bio *bio, struct btrfs_io_context *bioc);
void raid56_submit_missing_rbio(struct btrfs_raid_bio *rbio);

int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info);
void btrfs_free_stripe_hash_table(struct btrfs_fs_info *info);

#endif